| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters and Responses |
As they point out, it was not evident that we should include diacerein in this umbrella review of nonpharmacological and nonsurgical interventions for hip osteoarthritis (OA). Diacerein is a prescribed drug in several countries; however, in the majority of the European countries, including the Nordic region, this is not the case—and we thus chose to include it in this review. Diacerein is a diacetylated derivative of "rhein," which is the common name that describes the anthraquinone present in medicinal rhubarb (Rheirhizoma). This plant (the root) has been used in China as a medicinal herb for thousands of years, and thus provides the background for using the term "herbal therapy" to describe diacerein. We agree that this term is not entirely precise, given the different status of the substance in various countries.
Our umbrella review is based only on high-quality systematic reviews; primary studies were not considered in this process. As stated in our overview, "Adequate quantitative pooling of data in reviews was regarded as more valid than a qualitative data synthesis approach."1(p1720) We included only the part of the reviews with a quantitative pooling of patients who were diagnosed with hip OA. Thus, our analysis was based on the separate quantitative pooling for hip OA by Fidelix et al.2 We agree that it is not apparent that the information regarding included primary studies in Table 11(p1719) is based on 2 studies that address different outcomes; this could have been clarified by organizing the table differently.
Based on the separate quantitative pooling for hip OA by Fidelix et al (Table 2),2(p18) it appears that there might be a possible beneficial effect of diacerein on radiographic progression, with a relative risk of 0.84 (statistically significant); however, the absolute risk difference of –0.06 is not statistically significant. The review indicated marginal relative risk as well as inconsistent significance; therefore, we graded the results as low-quality evidence.
Addressing the question on why we graded the effect of diacerein on pain, impairment, or incidence of total hip replacement as moderate-quality evidence for no effect: we analyzed the quantitative pooling for hip OA of Fidelix et al2 separately. The results showed no statistically significant difference in favor of diacerein on hip OA; heterogeneity was taken into account. We graded it as moderate-quality evidence, referring to Table 2: Grading Quality of Evidence1(p1721); the results are based on 1 updated systematic review of high quality that is based on at least 1 high-quality primary study.
As Leeb and Rintelen point out, we are aware that Fidelix et al concluded that there is "gold"-level evidence that diacerein has a small, consistent benefit in pain improvement. However, this conclusion, as well as the effect estimates in the review by Rintelen et al,3 were based on effect estimates on both hip and knee OA, and not on a separate quantitative pooling for hip OA. Our conclusion is based solely on available data for hip OA.
We think that one of the most important goals of an umbrella review is identifying items where more research is needed, and the effectiveness of diacerein on hip OA is one of the areas detected by this method.
 | Footnotes |
|---|
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
